fix(cws): register connected sockets in the `flow_pid` map by YoannGh · Pull Request #54691 · DataDog/datadog-agent

YoannGh · 2026-08-10T22:39:08Z

What does this PR do?

Registers the flow of a connecting IPv6 socket once connect returns.

Motivation

tcp_v6_connect classifies the flow before the ephemeral source port and the source address are picked, so until Linux 7.0 the socket was classified on its first transmit, once both were known. Linux 7.0 only routes on a dst cache miss in inet6_csk_xmit, and a connecting socket always ends up holding a route, so that second classification never happens and those flows are left unattributed.

Describe how you validated your changes

Existing functional tests running on Ubuntu 26.04 that this PR fixes.

Additional Notes

IPv4 flows are unaffected as these are still registered as part of the security_sk_classify_flow hook.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 76e65fb118

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

datadog-official · 2026-08-10T22:54:29Z

🎯 Code Coverage (details)
• Patch Coverage: 0.00%
• Overall Coverage: 52.34% (+0.03%)

_{This comment will be updated automatically if new data arrives.

🔗 Commit SHA: 1af46af | Docs | Datadog PR Page | Give us feedback!}

dd-octo-sts · 2026-08-10T23:27:06Z

Files inventory check summary

File checks results against ancestor 5038adbf:

Results for datadog-agent_7.84.0~devel.git.146.1af46af.pipeline.130509364-1_amd64.deb:

No change detected

Results for datadog-iot-agent_7.84.0~devel.git.146.1af46af.pipeline.130509364-1_amd64.deb:

No change detected

dd-octo-sts · 2026-08-11T00:05:40Z

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor 5038adb
📊 Static Quality Gates Dashboard
🔗 SQG Job

Successful checks

Info

	Quality gate	Change	Size (prev → curr → max)
✅	agent_deb_amd64	+142.96 KiB (0.02% increase, -3.04% of buffer)	759.913 → 760.053 → 764.500
✅	agent_deb_amd64_fips	+142.96 KiB (0.02% increase, -4.71% of buffer)	712.777 → 712.917 → 715.740
✅	agent_rpm_amd64	+142.96 KiB (0.02% increase, -3.05% of buffer)	759.897 → 760.036 → 764.470
✅	agent_rpm_amd64_fips	+142.96 KiB (0.02% increase, -4.69% of buffer)	712.761 → 712.901 → 715.740
✅	agent_rpm_arm64	+141.56 KiB (0.02% increase, -18.72% of buffer)	735.802 → 735.940 → 736.540
✅	agent_rpm_arm64_fips	+141.56 KiB (0.02% increase, -8.14% of buffer)	691.992 → 692.130 → 693.690
✅	agent_suse_amd64	+142.96 KiB (0.02% increase, -3.05% of buffer)	759.897 → 760.036 → 764.470
✅	agent_suse_amd64_fips	+142.96 KiB (0.02% increase, -4.69% of buffer)	712.761 → 712.901 → 715.740
✅	agent_suse_arm64	+141.56 KiB (0.02% increase, -18.72% of buffer)	735.802 → 735.940 → 736.540
✅	agent_suse_arm64_fips	+141.56 KiB (0.02% increase, -8.14% of buffer)	691.992 → 692.130 → 693.690
✅	docker_agent_amd64	+142.96 KiB (0.02% increase, -9.21% of buffer)	818.525 → 818.664 → 820.040
✅	docker_agent_arm64	+141.56 KiB (0.02% increase, -10.95% of buffer)	819.477 → 819.615 → 820.740
✅	docker_agent_jmx_amd64	+142.96 KiB (0.01% increase, -10.13% of buffer)	1009.422 → 1009.562 → 1010.800
✅	docker_agent_jmx_arm64	+141.56 KiB (0.01% increase, -9.92% of buffer)	999.027 → 999.165 → 1000.420

19 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_heroku_amd64	312.396 MiB
✅	agent_msi	643.509 MiB
✅	docker_cluster_agent_amd64	210.475 MiB
✅	docker_cluster_agent_arm64	223.455 MiB
✅	docker_cws_instrumentation_amd64	7.439 MiB
✅	docker_cws_instrumentation_arm64	6.877 MiB
✅	docker_dogstatsd_amd64	39.482 MiB
✅	docker_dogstatsd_arm64	37.560 MiB
✅	docker_host_profiler_amd64	305.816 MiB
✅	docker_host_profiler_arm64	317.125 MiB
✅	dogstatsd_deb_amd64	30.224 MiB
✅	dogstatsd_deb_arm64	28.247 MiB
✅	dogstatsd_rpm_amd64	30.224 MiB
✅	dogstatsd_suse_amd64	30.224 MiB
✅	iot_agent_deb_amd64	46.448 MiB
✅	iot_agent_deb_arm64	43.104 MiB
✅	iot_agent_deb_armhf	43.887 MiB
✅	iot_agent_rpm_amd64	46.448 MiB
✅	iot_agent_suse_amd64	46.447 MiB

cit-pr-commenter-54b7da · 2026-08-11T00:19:12Z

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 8089dbeb-e6c7-4f47-95b3-bc5475f3b2a7

Baseline: 645110a
Comparison: 7a0dfc5
Diff

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_private_action_runner	memory utilization	+0.22	[+0.09, +0.35]	1	Logs bounds checks dashboard
➖	quality_gate_security_idle	memory utilization	+0.22	[+0.12, +0.32]	1	Logs bounds checks dashboard
➖	quality_gate_idle	memory utilization	+0.14	[+0.02, +0.26]	1	Logs bounds checks dashboard
➖	quality_gate_metrics_logs	memory utilization	+0.02	[-0.22, +0.27]	1	Logs bounds checks dashboard
➖	quality_gate_logs	% cpu utilization	+0.01	[-0.83, +0.85]	1	Logs bounds checks dashboard
➖	quality_gate_idle_all_features	memory utilization	-0.01	[-0.05, +0.03]	1	Logs bounds checks dashboard
➖	quality_gate_security_no_fs_load	memory utilization	-0.26	[-0.41, -0.12]	1	Logs bounds checks dashboard
➖	quality_gate_security_mean_fs_load	memory utilization	-0.28	[-0.35, -0.20]	1	Logs bounds checks dashboard

Bounds Checks: ❌ Failed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	quality_gate_idle	intake_connections	10/10	4 = 4	bounds checks dashboard
❌	quality_gate_idle	memory_usage	0/10	172.40MiB > 154MiB	bounds checks dashboard
✅	quality_gate_idle	total_bytes_received	10/10	729.33KiB ≤ 819.20KiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	4 = 4	bounds checks dashboard
❌	quality_gate_idle_all_features	memory_usage	8/10	516.58MiB > 512MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	total_bytes_received	10/10	1.12MiB ≤ 1.25MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	18 ≤ 40	bounds checks dashboard
❌	quality_gate_logs	memory_usage	0/10	205.74MiB > 195MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_logs	total_bytes_received	10/10	264.20MiB ≤ 292MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	424.55 ≤ 2000	bounds checks dashboard
✅	quality_gate_metrics_logs	intake_connections	10/10	15 ≤ 40	bounds checks dashboard
✅	quality_gate_metrics_logs	memory_usage	10/10	410.53MiB ≤ 430MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	total_bytes_received	10/10	0.94GiB ≤ 1.04GiB	bounds checks dashboard
✅	quality_gate_private_action_runner	memory_usage	10/10	72.38MiB ≤ 75MiB	bounds checks dashboard
✅	quality_gate_security_idle	cpu_usage	10/10	31.62 ≤ 100	bounds checks dashboard
✅	quality_gate_security_idle	memory_usage	10/10	327.00MiB ≤ 330MiB	bounds checks dashboard
✅	quality_gate_security_mean_fs_load	cpu_usage	10/10	59.73 ≤ 200	bounds checks dashboard
✅	quality_gate_security_mean_fs_load	memory_usage	10/10	302.20MiB ≤ 310MiB	bounds checks dashboard
✅	quality_gate_security_no_fs_load	cpu_usage	10/10	20.16 ≤ 100	bounds checks dashboard
✅	quality_gate_security_no_fs_load	memory_usage	10/10	312.26MiB ≤ 320MiB	bounds checks dashboard

Explanation

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

Performance changes are noted in the perf column of each table:

✅ = significantly better comparison variant performance
❌ = significantly worse comparison variant performance
➖ = no significant change in performance

A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".

For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:

Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.
Its configuration does not mark it "erratic".

Replicate Execution Details

We run multiple replicates for each experiment/variant. However, we allow replicates to be automatically retried if there are any failures, up to 8 times, at which point the replicate is marked dead and we are unable to run analysis for the entire experiment. We call each of these attempts at running replicates a replicate execution. This section lists all replicate executions that failed due to the target crashing or being oom killed.

Note: In the below tables we bucket failures by experiment, variant, and failure type. For each of these buckets we list out the replicate indexes that failed with an annotation signifying how many times said replicate failed with the given failure mode. In the below example the baseline variant of the experiment named experiment_with_failures had two replicates that failed by oom kills. Replicate 0, which failed 8 executions, and replicate 1 which failed 6 executions, all with the same failure mode.

Experiment	Variant	Replicates	Failure	Logs	Debug Dashboard
experiment_with_failures	baseline	0 (x8) 1 (x6)	Oom killed		Debug Dashboard

The debug dashboard links will take you to a debugging dashboard specifically designed to investigate replicate execution failures.

❌ Retried Profiling Replicate Execution Failures (ddprof)

Note: Profiling replicas may still be executing. See the debug dashboard for up to date status.

Experiment	Variant	Replicates	Failure	Debug Dashboard
quality_gate_idle	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	baseline	10	Oom killed	Debug Dashboard
quality_gate_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_logs	comparison	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	comparison	10	Oom killed	Debug Dashboard
quality_gate_security_idle	baseline	10	Crashed (exit code: 134)	Debug Dashboard
quality_gate_security_idle	comparison	10	Oom killed	Debug Dashboard
quality_gate_security_mean_fs_load	baseline	10	Crashed (exit code: 134)	Debug Dashboard
quality_gate_security_mean_fs_load	comparison	10	Crashed (exit code: 134)	Debug Dashboard
quality_gate_security_no_fs_load	comparison	10	Crashed (exit code: 134)	Debug Dashboard

CI Pass/Fail Decision

❌ Failed. Some Quality Gates were violated.

quality_gate_metrics_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
quality_gate_metrics_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
quality_gate_metrics_logs, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
quality_gate_metrics_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
quality_gate_metrics_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
quality_gate_idle_all_features, bounds check memory_usage: 8/10 replicas passed. Failed 2 which is > 0. Gate FAILED.
quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
quality_gate_idle_all_features, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
quality_gate_security_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
quality_gate_security_idle, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
quality_gate_security_no_fs_load, bounds check memory_usage: 10/10 replicas passed. Gate passed.
quality_gate_security_no_fs_load, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
quality_gate_private_action_runner, bounds check memory_usage: 10/10 replicas passed. Gate passed.
quality_gate_idle, bounds check memory_usage: 0/10 replicas passed. Failed 10 which is > 0. Gate FAILED.
quality_gate_idle, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
quality_gate_security_mean_fs_load, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
quality_gate_security_mean_fs_load, bounds check memory_usage: 10/10 replicas passed. Gate passed.
quality_gate_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
quality_gate_logs, bounds check memory_usage: 0/10 replicas passed. Failed 10 which is > 0. Gate FAILED.
quality_gate_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.

dd-octo-sts · 2026-08-12T13:54:57Z

eBPF complexity changes

Summary result: ✅ - stable

Highest complexity change (%): +2.67%
Highest complexity change (abs.): +2 instructions
Programs that were above the 85.0% limit of instructions and are now below: 0
Programs that were below the 85.0% limit of instructions and are now above: 0

Groups without changes:

btf_test
dns
ebpf
error_telemetry
gpu
offset_guess
runtime_security_fentry
runtime_security_offset_guesser
shared_libraries
sleepable
tracer
tracer_fentry
uprobe_attacher_test
usm
usm_events_test

runtime_security details

runtime_security [programs with changes]

Program	Avg. complexity	Distro with highest complexity	Distro with lowest complexity
hook_security_socket_connect	🔴 77.5 (+2.0, +2.65%)	ubuntu_18.04/x86_64: 🔴 78.0 (+2.0, +2.63%)	debian_10/x86_64: 🔴 77.0 (+2.0, +2.67%)

runtime_security_syscall_wrapper details

runtime_security_syscall_wrapper [programs with changes]

Program	Avg. complexity	Distro with highest complexity	Distro with lowest complexity
hook_security_socket_connect	🔴 77.5 (+2.0, +2.65%)	ubuntu_18.04/x86_64: 🔴 78.0 (+2.0, +2.63%)	debian_10/x86_64: 🔴 77.0 (+2.0, +2.67%)

This report was generated based on the complexity data for the current branch yoanngh/connect-flow-pid-kernel-7 (pipeline 130509364, commit 1af46af) and the base branch main (commit 5038adb). Objects without changes are summarized by group only. Contact #ebpf-platform if you have any questions/feedback.

Table complexity legend: 🔵 - new; ⚪ - unchanged; 🟢 - reduced; 🔴 - increased

theop-dd · 2026-08-12T14:23:45Z

+// Before Linux 7.0 an IPv6 socket was classified on its first transmit, once its source address and port
+// were known. Starting with Linux 7.0 inet6_csk_xmit only calls security_sk_classify_flow on a route miss.
+__attribute__((always_inline)) int register_native_ipv6_flow(struct sock *sk, u64 pid_tgid) {
+    // IPv4 and ipv4 mapped addresses still reach security_sk_classify_flow with a usable flow


repetition I assume?

The comment is correct but not really clear: this function does not need to do anything for two different cases:

for a socket using IPv4

for a socket using IPv6 but using IPv4 addresses

I split the comment into two parts to make that more clear

security_sk_classify_flow runs in tcp_v6_connect before the ephemeral source port and the source address have been picked, so the flow it reports for a connecting socket is incomplete. Until now the transmit path called the hook again once both were known; since Linux 7.0 inet6_csk_xmit only routes on a dst cache miss, so that second call never happens and IPv6 TCP client flows are either registered under a wildcard address or, when the socket was never bound, not registered at all. Register the flow on the connect syscall exit instead, reading the final port and source address from the socket. The connect probes are now loaded along with the network probes so that flow_pid stays up to date even when no rule asks for connect events.

YoannGh requested a review from a team as a code owner August 10, 2026 22:39

YoannGh added the changelog/no-changelog No changelog entry needed label Aug 10, 2026

YoannGh changed the title ~~fix(cws): register connected sockets in the flow to PID map~~ fix(cws): register connected sockets in the flow_pid map Aug 10, 2026

chatgpt-codex-connector Bot reviewed Aug 10, 2026

View reviewed changes

Comment thread pkg/security/ebpf/c/include/hooks/network/flow.h Outdated

Comment thread pkg/security/ebpf/c/include/maps.h Outdated

github-actions Bot added the medium review PR review might take time label Aug 10, 2026

YoannGh force-pushed the yoanngh/connect-flow-pid-kernel-7 branch from 76e65fb to 75a2484 Compare August 10, 2026 23:21

Base automatically changed from yoanngh/accept-flow-pid-kernel-7 to main August 11, 2026 09:14

YoannGh added team/agent-security qa/done QA done before merge and regressions are covered by tests category/bugfix labels Aug 11, 2026

YoannGh force-pushed the yoanngh/connect-flow-pid-kernel-7 branch from 75a2484 to 68f8c7d Compare August 11, 2026 10:28

github-actions Bot added the component/system-probe label Aug 11, 2026

dd-octo-sts Bot added the internal Identify a non-fork PR label Aug 11, 2026

YoannGh force-pushed the yoanngh/connect-flow-pid-kernel-7 branch 2 times, most recently from b687a36 to e650c1b Compare August 12, 2026 12:39

theop-dd reviewed Aug 12, 2026

View reviewed changes

theop-dd approved these changes Aug 12, 2026

View reviewed changes

YoannGh added 2 commits August 12, 2026 17:29

simplify and split comment for register_native_ipv6_flow

1af46af

YoannGh force-pushed the yoanngh/connect-flow-pid-kernel-7 branch from 6dc955e to 1af46af Compare August 12, 2026 15:30

gh-worker-dd-mergequeue-cf854d Bot merged commit 7a0dfc5 into main Aug 12, 2026
440 checks passed

gh-worker-dd-mergequeue-cf854d Bot deleted the yoanngh/connect-flow-pid-kernel-7 branch August 12, 2026 18:26

github-actions Bot added this to the 7.84.0 milestone Aug 12, 2026

Navigation Menu

Uh oh!

fix(cws): register connected sockets in the `flow_pid` map - #54691

fix(cws): register connected sockets in the `flow_pid` map#54691
gh-worker-dd-mergequeue-cf854d[bot] merged 2 commits into
mainfrom
yoanngh/connect-flow-pid-kernel-7

YoannGh commented Aug 10, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

Uh oh!

Uh oh!

datadog-official Bot commented Aug 10, 2026 •

edited by datadog-prod-us1-3 Bot

Loading

Uh oh!

dd-octo-sts Bot commented Aug 10, 2026 •

edited

Loading

Uh oh!

dd-octo-sts Bot commented Aug 11, 2026 •

edited

Loading

Info

Uh oh!

cit-pr-commenter-54b7da Bot commented Aug 11, 2026 •

edited

Loading

Fine details of change detection per experiment

Explanation

Replicate Execution Details

❌ Retried Profiling Replicate Execution Failures (ddprof)

Uh oh!

dd-octo-sts Bot commented Aug 12, 2026 •

edited

Loading

runtime_security [programs with changes]

runtime_security_syscall_wrapper [programs with changes]

Uh oh!

theop-dd Aug 12, 2026

Uh oh!

YoannGh Aug 12, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

YoannGh commented Aug 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What does this PR do?

Motivation

Describe how you validated your changes

Additional Notes

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

datadog-official Bot commented Aug 10, 2026 • edited by datadog-prod-us1-3 Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dd-octo-sts Bot commented Aug 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Files inventory check summary

Results for datadog-agent_7.84.0~devel.git.146.1af46af.pipeline.130509364-1_amd64.deb:

Results for datadog-iot-agent_7.84.0~devel.git.146.1af46af.pipeline.130509364-1_amd64.deb:

Uh oh!

dd-octo-sts Bot commented Aug 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Static quality checks

Info

Uh oh!

cit-pr-commenter-54b7da Bot commented Aug 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Regression Detector

Regression Detector Results

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment

Bounds Checks: ❌ Failed

Explanation

Replicate Execution Details

❌ Retried Profiling Replicate Execution Failures (ddprof)

CI Pass/Fail Decision

Uh oh!

dd-octo-sts Bot commented Aug 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

eBPF complexity changes

Summary result: ✅ - stable

runtime_security [programs with changes]

runtime_security_syscall_wrapper [programs with changes]

Uh oh!

theop-dd Aug 12, 2026

Choose a reason for hiding this comment

Uh oh!

YoannGh Aug 12, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

YoannGh commented Aug 10, 2026 •

edited

Loading

datadog-official Bot commented Aug 10, 2026 •

edited by datadog-prod-us1-3 Bot

Loading

dd-octo-sts Bot commented Aug 10, 2026 •

edited

Loading

dd-octo-sts Bot commented Aug 11, 2026 •

edited

Loading

cit-pr-commenter-54b7da Bot commented Aug 11, 2026 •

edited

Loading

dd-octo-sts Bot commented Aug 12, 2026 •

edited

Loading